← Back to community benchmarks
Laguna-S-2.1-8bit-Att-6bit-Ex
Performance
64k
tokens
730.9
PP tok/s
38.3
TG tok/s
89663
TTFT (ms)
93.8
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.5.3
macOS
macOS 26.5.2
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,002 | 54.3 | 90.8 GB | view |
| 4k | 981.0 | 53.2 | 90.9 GB | view |
| 8k | 922.5 | 51.8 | 91.1 GB | view |
| 16k | 884.4 | 49.5 | 91.5 GB | view |
| 32k | 778.7 | 44.8 | 92.2 GB | view |
| 64k | 730.9 | 38.3 | 93.8 GB | current |
| 128k | 619.5 | 30.8 | 97.1 GB | view |
| 195k | 523.9 | 25.4 | 100.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 54.3 | 1.00× |
| 2× | 112.9 | 2.08× |
| 4× | 225.2 | 4.15× |
| 8× | 449.7 | 8.28× |